Quote from: weeminer on January 29, 2018, 03:59:10 PM

Quote from: siforek on January 26, 2018, 09:23:46 PM

Quote from: dstm on January 25, 2018, 03:05:20 PM

Quote from: siforek on January 24, 2018, 01:03:25 AM

Quote from: betopica on January 24, 2018, 12:17:59 AM

Quote from: siforek on January 23, 2018, 10:41:35 PM

I tried switching from ewbf to dstm for the 2nd time yesterday. Initially it started out 5-10% faster than ewbf which is awesome, however after about 10-15 minutes it slowed more & more to the point where it was as much as 40% slower than ewbf, at which point I switched back.
I've talked to others that do not have the same problem, yet they were running less GPUs than I am. I have multiple rigs with 13x1070s & 1070tis.

I have the same problem with DSTM 0.5.6, 0.5.7 and 0.5.8. Stable and fast as hell, but after one hour it starts to make less and less sol/s.. in three hours I'm getting half rate (I start with 680-700 sol/s and after 4 hours I end with 230 sol/s)... so my only solution is restart the miner. All the time no more than 60ºC, fan speed at 60%. I have two rigs, same specs and the same problem: Asrock H110 PRO + 13 Nvidia 1080Ti (up five different brands, mixed). Every rig have 4 PSUs: 3 x 1000W + 1 x 1600W and just one cable to power every riser (4 pin molex version) so no problems about power or temperature. I use simplemining without overclock, just power at 200W, no extra mhz in gpu core or memory. Is it a configuration problem?

Yep same motherboard, never had this issue with any other miner but I'd love to get it working & get the extra sols. Anyone else have any ideas?

Are you all using SMOS?
Some SMOS user reported something similar recently.
https://bitcointalk.org/index.php?topic=2021765.msg27398311#msg27398311

If you take a look at the provided data: affected systems seems to have a lot 'nvidia-settings' processes running, this might be the reason for the performance drop after some time. So pls check if this is also the case on your system - It looks like SMOS is starting nvidia-setting (most likely for monitoring) without properly closing it.

Having this issue & yes, on SMOS. Would just replace SMOS with Ubuntu if I could but only have remote access to rigs. Might need to mod SMOS source to fix :/

I am also having this issue..running with smos. With 12 x gtx1060..rig started with 3700sol/s..after running for 2 hours the hash rate drop to 2000sol/s. Is there anyway we can fix this in smos ? I tried both 0.5.7 and 0.5.8 same result..Now trying 0.5.6 version

I guys..I am running v0.5.6 in smos with 12x gtx1060..Somehow manage to get it stable running around 3700sol/s for 12hours..No decrease in hashrate so far...I assume it's stable right now compare to v0.5.7 and v0.5.8 in smos..this might be the resolution for those having similar issue